SIG-0321
- LLMs
Google Confirms Gemini AI Breached Three Firms
Google is the latest AI giant to confirm that its models escaped a testing environment and hacked real companies.
via SecurityWeekOpen
AI Security
No signal on this page matches that topic.
SIG-0322
- AI-powered attacks
A week in security (September 14 – September 20)
Last week on Malwarebytes Labs: New Android malware uses AI to steal bank logins and PINs Did an AI really try to break…
via Malwarebytes LabsOpen
SIG-0323
- Defence & detection
Product showcase: Helmit alerts parents when online conversations show signs of trouble
Helmit is a parental control app that combines AI-powered social media monitoring with screen time management, web…
via Help Net SecurityOpen
SIG-0324
- Policy & governance
Intent injection attacks are a new worry for AI-native 6G networks
Intent-based networking (IBN) lets operators state the outcome they want and leaves its translation into network policy to software, an approach AI-native 6G designs have moved to the forefront…
via Help Net SecurityOpen
SIG-0325
- Policy & governance
AI compliance issues hit 2 in 5 large companies, and legacy workflows are a big factor
Forty percent of large companies had an AI-related compliance or governance issue in the past 12 months, according to…
via Help Net SecurityOpen
SIG-0326
- Privacy & data
The Right Tool for the Job: On the Selection of Mitigations for GenAI Privacy Threats
arXiv:2609.20884v1 Announce Type: new Abstract: Generative Artificial Intelligence (GenAI) has rapidly evolved from an…
via arXiv cs.CROpen
SIG-0327
- AI security
TPM-Attest: Hardware-Rooted Integrity Attestation as a Kernel-Level Anti-Cheat Alternative for Linux
arXiv:2609.20909v1 Announce Type: new Abstract: Multiplayer PC gaming on Linux faces a structural problem: the…
via arXiv cs.CROpen
SIG-0328
- AI security
(Don't) Trust, but (Don't) Verify: Developers' Attention to Security in AI-Generated Code
arXiv:2609.21020v1 Announce Type: new Abstract: AI coding assistants are rapidly transforming software development, but…
via arXiv cs.CROpen
SIG-0329
- AI security
Loopjacking: Hijacking Human-in-the-Loop Approval
arXiv:2609.21081v1 Announce Type: new Abstract: Human approval is often treated as the last security boundary before an agent executes a consequential operation. That boundary is only meaningful if…
via arXiv cs.CROpen
SIG-0330
- Prompt injection
- LLMs
Origin Is All You Need: Provenance-Aware Transformers for Structural Trust-Boundary Separation
arXiv:2609.21088v1 Announce Type: new Abstract: Indirect prompt injection (IPI) remains a central safety and security…
via arXiv cs.CROpen
SIG-0331
- Policy & governance
- LLMs
NetInspector: Measuring and Improving LLM Capabilities for Reliable Intent-Based Networking Policy Generation
arXiv:2609.21103v1 Announce Type: new Abstract: Modern networks are large in scale and heterogeneous in configuration…
via arXiv cs.CROpen
SIG-0332
- Defence & detection
X-SPUR: Explainable Surprisal-Based Protocol-Aware Unsupervised Reasoning for Automotive Ethernet Intrusion Detection
arXiv:2609.21217v1 Announce Type: new Abstract: Automotive Ethernet carries heterogeneous multi-protocol traffic in…
via arXiv cs.CROpen
SIG-0333
- AI security
Transcript-Bound Combiners for Downgrade-Resilient Hybrid Post-Quantum Key Establishment: Definition, Proof, and Embedded-Device Cost
arXiv:2609.21273v1 Announce Type: new Abstract: Hybrid key establishment runs a post-quantum key-encapsulation…
via arXiv cs.CROpen
SIG-0334
- AI-powered attacks
Identifying Security Platform Product Abuse with Machine Learning
arXiv:2609.21303v1 Announce Type: new Abstract: Product abuse is an individually rare, but growing, problem across the SaaS industry. Highly sophisticated threat actors can misuse security platforms…
via arXiv cs.CROpen
SIG-0335
- Privacy & data
- Policy & governance
Conformal Privacy Auditing: Calibrated Re-identification Attacks with Statistical Guarantees
arXiv:2609.21340v1 Announce Type: new Abstract: Empirical identity leakage from released text is increasingly driven by…
via arXiv cs.CROpen
SIG-0336
- LLMs
CESBench: Benchmarking Large Language Models on Cryptographic Engineering Security for IoT Devices
arXiv:2609.21344v1 Announce Type: new Abstract: For Internet of Things (IoT) devices, a secure algorithm alone is not…
via arXiv cs.CROpen
SIG-0337
- Prompt injection
- Privacy & data
HE-Guardrail: A Homomorphic Guardrail Against Jailbreak Attacks for Encrypted Large Language Model Inference
arXiv:2609.21484v1 Announce Type: new Abstract: Homomorphic encryption (HE) has emerged as a promising approach to…
via arXiv cs.CROpen
SIG-0338
- Model attacks
- LLMs
Micro-Collaborative Poisoning: A Distributed Attack on RAG Systems
arXiv:2609.21573v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) improves large language models by…
via arXiv cs.CROpen
SIG-0339
- Agents
- Privacy & data
CIPL: A Channel-Aware Framework for Recoverable Privacy Leakage in LLM Agents
arXiv:2609.21686v1 Announce Type: new Abstract: Privacy leakage in LLM agents is commonly evaluated within individual components such as memory, retrieval, or tool-use pipelines, which makes it…
via arXiv cs.CROpen
SIG-0340
- Model attacks
TERMon: Detecting Persistent Behavioral Threats in Edge AI via Hardware-Native Ternary Runtime Monitor
arXiv:2609.21713v1 Announce Type: new Abstract: Edge AI accelerators are increasingly deployed in safety-critical…
via arXiv cs.CROpen